Journal Article

High-precision and linear weight updates by subnanosecond pulses in ferroelectric tunnel junction for neuro-inspired computing

Yin, Yuewei, Luo, Zhen, Guan, Zeyu, Wang, He, Wang, Zijian, Ma, Chao, Li, Xiaoguang, Zhao, Letian, Sun, Haoyang, Jin, Xi, Liu, Chuanchuan, Lin, Yue

2022

Overview

The rapid development of neuro-inspired computing demands synaptic devices with ultrafast speed, low power consumption, and multiple non-volatile states, among other features. Here, a high-performance synaptic device is designed and established based on a Ag/PbZr 0.52 Ti 0.48 O 3 (PZT, (111)-oriented)/Nb:SrTiO 3 ferroelectric tunnel junction (FTJ). The advantages of (111)-oriented PZT (~1.2 nm) include its multiple ferroelectric switching dynamics, ultrafine ferroelectric domains, and small coercive voltage. The FTJ shows high-precision (256 states, 8 bits), reproducible (cycle-to-cycle variation, ~2.06%), linear (nonlinearity <1) and symmetric weight updates, with a good endurance of >10 9 cycles and an ultralow write energy consumption. In particular, manipulations among 150 states are realized under subnanosecond (~630 ps) pulse voltages ≤5 V, and the fastest resistance switching at 300 ps for the FTJs is achieved by voltages <13 V. Based on the experimental performance, the convolutional neural network simulation achieves a high online learning accuracy of ~94.7% for recognizing fashion product images, close to the calculated result of ~95.6% by floating-point-based convolutional neural network software. Interestingly, the FTJ-based neural network is very robust to input image noise, showing potential for practical applications. This work represents an important improvement in FTJs towards building neuro-inspired computing systems. Brain-inspired computing demands high-performance synapses. Here, the authors report a subnanosecond ferroelectric tunnel junction with 256 conductance states, 10 9 endurance, and 5.3 fJ/bit energy consumption, satisfactory to build synaptic devices.

• Publisher: Nature Publishing Group UK,Nature Publishing Group,Nature Portfolio
• Subject: 147/135 / 147/136 / 147/137 / 147/3 / 639/301/1005/1007 / 639/301/357 / 639/766/119/996 / Artificial neural networks / Computation / Endurance / Energy consumption / Ferroelectric domains / Ferroelectric materials / Ferroelectricity / Floating point arithmetic / Humanities and Social Sciences / Machine learning / multidisciplinary / Nanosecond pulses / Neural networks / Nonlinear systems / Object recognition / Power consumption / Science / Science (multidisciplinary) / Switching / Synapses / Tunnel junctions / Ultrafines